Identification of the Missing Protein Hyaluronan Synthase 1 in Human

1 Proteomics Unit. Faculty of Pharmacy, Complutense University of Madrid (UCM). Plaza ... Madrid, Spain. 3 Andalusian Public Health System Biobank. Av...
3 downloads 0 Views 497KB Size
Download PDF
Letter pubs.acs.org/jpr

Cite This: J. Proteome Res. XXXX, XXX, XXX−XXX

Identification of the Missing Protein Hyaluronan Synthase 1 in Human Mesenchymal Stem Cells Derived from Adipose Tissue or Umbilical Cord Luis Felipe Clemente,† María Luisa Hernáez,† Antonio Ramos-Fernández,‡ Gertrudis Ligero,§ Concha Gil,†,∥ Fernando José Corrales,⊥,# and Miguel Marcilla*,⊥,#,▽ †

Proteomics Unit, Faculty of Pharmacy, Complutense University of Madrid (UCM), Plaza Ramón y Cajal s/n, 28040 Madrid, Spain Proteobotics SL, Darwin 3, 28049 Madrid, Spain § Andalusian Public Health System Biobank, Avenida Del Conocimiento s/n, 18016 Granada, Spain ∥ Department of Microbiology & Parasitology, Faculty of Pharmacy, Complutense University of Madrid (UCM) and Ramón y Cajal Institute of Health Research (IRYCIS), Plaza Ramón y Cajal s/n, 28040 Madrid, Spain ⊥ Proteomics Unit, Spanish National Biotechnology Centre (CNB-CISC), Darwin 3, 28049 Madrid, Spain

Downloaded via KAOHSIUNG MEDICAL UNIV on July 8, 2018 at 11:31:05 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Currently, 14% of the human proteome is made up of proteins whose existence is not confirmed by mass spectrometry. We performed a proteomic profiling of human mesenchymal stem cells derived from adipose tissue or umbilical cord (PRIDE accession number: PXD009893) and identified peptides derived from 13 of such missing proteins. Remarkably, we found compelling evidence of the expression of hyaluronan synthase 1 (NX_Q92839-1) and confirmed its identification by the fragmentation of four heavy-labeled peptides that coeluted with their endogenous light counterparts. Our data also suggest that mesenchymal stem cells constitute a promising source for the detection of missing proteins.

■

■

INTRODUCTION

A detailed description of the methods employed can be found in Supporting Information S1. In brief, we lysed about 107 hMSCs derived from adipose tissue or umbilical cord. After precipitation, we subjected the proteins to two different analytical workflows. First, we fractionated the proteins by SDS-PAGE and performed an in-gel digestion with trypsin of the resulting 14 gel slices. After peptide extraction, we analyzed the samples by LC−MS/MS using a Q Exactive HF mass spectrometer (Thermo Fisher). In the second approach, we performed a trypsin digestion, and peptides were resolved by high-pH reversed-phase chromatography. We collected six chromatographic fractions that, together with an aliquot of the unfractionated sample, were analyzed by LC−MS/MS in a 5600 Triple-TOF mass spectrometer (Sciex). We searched the acquired MS/MS spectra against the last release available of the neXtProt database (v2.15.0; 01-172018) using an in-house developed algorithm that combines

Human mesenchymal stem cells (hMSCs) are multipotent stromal cells that can be isolated from a number of tissues and organs such as umbilical cord, bone marrow, liver, muscle, adipose tissue, or peripheral blood.1 Because of their wellestablished capacity for tissue repair, they are employed in regenerative medicine, mainly in clinical trials. In addition, they show great promise for the treatment of autoimmune diseases due to their unique immunomodulatory potential.2 For these reasons, there is growing interest in the study of hMSCs and, in particular, in the elucidation of the molecular mechanisms that determine their biological properties. We performed a large-scale shotgun analysis of these cells in the context of a research project involving the proteomic profiling of hMSCs derived from umbilical cord or adipose tissue. As hMSCs are undifferentiated cells,1 we reasoned that they could be a good source for the identification of missing proteins because they could express genes that are repressed in normal adult tissues. This Letter summarizes our findings in this regard and aims to highlight the potential of hMSCs for the detection of missing proteins. © XXXX American Chemical Society

MATERIALS AND METHODS

Special Issue: Human Proteome Project 2018 Received: May 29, 2018 Published: July 5, 2018 A

DOI: 10.1021/acs.jproteome.8b00384 J. Proteome Res. XXXX, XXX, XXX−XXX

Letter

Journal of Proteome Research Table 1. Protein Identifications after LC−MS/MS Analysis of hMSCs data set

FDRPSM (%)

FDRpept (%)

FDRprot (%)

PSMs

peptides

proteins

protein groups

adipose tissue (RP-QTOF) umbilical cord (RP-QTOF) adipose tissue (PAGE-Q Exactive) umbilical cord (PAGE-Q Exactive)

0.25 0.22 0.17 0.22

0.51 0.49 0.48 0.47

1.00 1.00 1.00 1.00

105 065 121 147 288 536 295 020

34 887 41 186 54 382 60 787

4616 5160 5943 6202

4166 4660 5274 5482

the outputs of four independent search engines:3,4 MASCOT,5 X! Tandem,6 OMSSA,7 and Myrimatch.8 Identifications were filtered at a FDR ≤ 1% at the protein level. It must be noted that FDR estimation is based on several imperfect assumptions, and thus not all proteins surpassing this threshold are confidently identified.

checked the status of these one-hit wonders in Peptide Atlas (Supporting Table S2) and found that (1) six of them lacked any experimental confirmation (status: nondetected), (2) in four cases, the evidence of their expression was weak (status: weak or marginally distinguished), and (3) the identification of two proteins was supported by two distinct tryptic peptides (status: representative or subsumed). With the information available in Peptide Atlas and the peptides longer than eight residues reported here, three of the proteins in the second group (NX_Q6TDP4-1, NX_Q3ZCT8-1, and NX_Q96HZ42) could be classified as nonmissing. Nevertheless, we consider that this claim should be validated by targeted analysis using labeled standards and are currently working in this direction. In addition to these 12 one-hit wonders, we noticed that hyaluronan synthase 1 (NX_Q92839-1) was detected in three of the four data sets (Figure 1). Furthermore, this identification was supported by six different peptide sequences (Table 2 and Supporting Figure S1). The expression of hyaluronan synthase 1 was particularly evident in the adipose tissue-derived hMSCs, where it was identified with five different tryptic peptides. Because four of the sequences in this set are of at least nine amino acids in length, the detection of this missing protein is validated according to the HPP data interpretation guidelines 2.1.10 However, for further confirmation, we set out to synthesize five of the identified peptides isotopically labeled with 13C6 15N4 arginine (+10 Da) at the C-terminus. We ruled out the longest one (ACQSYFHCVSCSGPLGLYR, 20 residues long) to avoid problems during solid-phase synthesis. These five heavy standards were spiked into tryptic digests of umbilical cord or adipose tissue-derived hMSCs, and the resultant mixture was analyzed by parallel reaction monitoring (PRM) in both the Q Exactive and the 5600 Triple TOF platforms. This analysis proved that with only one exception (LAVEALVR), all of the light peptides coeluted with their heavy counterparts (Supporting Figure S2). In agreement with the shotgun analysis, hyaluronan synthase 1 was identified with much more confidence in the extract of hMSCs derived from adipose tissue. Regarding this latter point, only one endogenous peptide (LDPMALLELVR) was detected in umbilical cord-derived hMSCs and exclusively in the analysis performed in the Q Exactive mass spectrometer (Supporting

■

RESULTS AND DISCUSSION We identified between 4616 and 6202 proteins depending on the origin of the hMSCs and the analytical workflow. Not surprisingly, the analysis performed in the Q Exactive platform after SDS-PAGE separation resulted in a 20% (umbilical cord) and 28% (adipose tissue) increase in protein matches, likely reflecting the higher performance of this mass spectrometer and the more extensive fractionation of the samples (Table 1). Next, we queried these lists of identifications for the presence of missing proteins. We used the neXtProt peptide uniqueness checker9 to determine which of the sequences assigned to missing proteins were unique, taking into account all known sequence variation arising from single amino acid substitutions. Thirteen candidates survived this stringent filtering (Figure 1 and Supporting Table S1). We gathered

Figure 1. Putative missing proteins identified in extracts of adipose tissue and umbilical-cord-derived hMSCs. The proteins in the upper row were detected (green squares) or not (red squares) after LC− MS/MS analysis of the samples described on the left column. The numbers inside the green squares indicate the number of unique peptides identified in each case.

evidence of the presence of 5 and 11 missing proteins in the adipose tissue and the umbilical cord samples, respectively. All in all, the overlap between the four data sets was low, and 12 out of the 13 matches relied on a single unique peptide. We

Table 2. Unique Peptides Mapped to Hyaluronan Synthase 1 (NX_Q92839-1) peptide

m/z

z

AT Triple TOFa

UC Triple TOFb

AT Q Exactivea

UC Q Exactiveb

confirmed

AcQSYFHcVScISGPLGLYR LDPMALLELVR GPLDAATAR YWVAFNVER ADWSGPSR LAVEALVR

792.36 635.37 436.24 592.30 438.20 435.77

3+ 2+ 2+ 2+ 2+ 2+

+ + + + + −

− − − − − +

− + + − − −

− − − − − −

not tested yes yes yes yes no

a

AT: adipose tissue. bUC: umbilical cord. B

DOI: 10.1021/acs.jproteome.8b00384 J. Proteome Res. XXXX, XXX, XXX−XXX

Letter

Journal of Proteome Research Figure S2). This observation is probably indicative of differential regulation of HAS1 expression between both cells. It is worth noting that one of the peptides identified in the shotgun analysis (LAVEALVR) failed to coelute with its labeled counterpart despite having very similar MS/MS spectra (Supporting Figure S3A,B). In our opinion, the most likely explanation to this fact is that the endogenous species corresponds to the peptide LAVEAVLR derived from the beta subunit of the CCT chaperone complex (NX_P78371). Both sequences are very similar and differ only in the order of residues 6 and 7, which accounts for the small shift in retention time. Furthermore, the alternative sequence matches better with the MS/MS spectrum, as this shows signals compatible with the product ions y2+, y2+-17, and b6+ (Supporting Figure S3C). Whatever the explanation, this example highlights the need to confirm peptide detection claims with SRM/PRM experiments using labeled standards and not only by fragmentation of the corresponding synthetic peptides. Thus we propose that this sort of verification should be mandatory for reporting the identification of missing proteins in the context of the Human Proteome Project. Hyaluronan synthase 1 (EC 2.4.1.212) is encoded by the HAS1 gene and belongs to a family of three isozymes that participate in the synthesis of hyaluronic acid, catalyzing the addition of N-acetylglucosamine or glucuronic acid to the nascent polymer.11 The identification of this enzyme in the proteome of hMSCs is not completely unexpected as mRNA levels of HAS1 are relatively high in bone-marrow-derived hMSCs.12 It is likely that this is also the case for cells of different origins because the hyaluronan receptor CD44 is a positive marker of stemness and the interaction of hMSCs with the extracellular matrix is critical for their physiological function.1 Considering its well established physiological function and its extensive annotation, the lack of MS-based evidence for hyaluronan synthase 1 in the literature seems surprising. It is possible that the localization of this protein at the cell membrane, dictated by the presence of seven transmembrane regions, poses a drawback to its efficient extraction and solubilization, hampering its identification by standard proteomic workflows. Anyhow, here we demonstrate the expression of HAS1 at the protein level so that it should no longer be considered a missing protein. Besides, the four peptides validated in this work are ideal targets for the development of PRM/MRM methods. Finally, we provide suggestive evidence of the presence of 12 additional missing proteins that deserve further analysis for their confirmation, indicating that hMSCs may constitute a good model to gain insight into this hidden portion of the human proteome.

■

■

Figure S2. PRM analysis of five tryptic peptides derived from hyaluronan synthase 1. Supporting Figure S3. Extracted ion chromatograms and MS/MS spectra of the peptides LAVEALVR[+10] and LAVEAVLR. (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: (+34) 91 5854540. Fax: (+34) 91 5854506. ORCID

Miguel Marcilla: 0000-0001-9171-5076 Present Address ▽

M.M.: Spanish National Biotechnology Centre, Darwin 3, 28049 Madrid, Spain.

Author Contributions #

F.J.C. and M.M. contributed equally.

Notes

The authors declare no competing financial interest. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http:// proteomecentral.proteomexchange.org) via the PRIDE partner repository with the data set identifier PXD009893.

■

ACKNOWLEDGMENTS The Proteomics Units of the Spanish National Biotechnology Centre and of the Complutense University of Madrid belong to Proteored-ISCIII and are supported by grant PT17/0019 of the PE I+D+i 2013-2016, funded by ISCIII and ERDF. The Andalusian Public Health System Biobank is part of the National Bank of Stem Cell Lines (BNLC-ISCIII) and is also supported by the above-mentioned grant.

■

REFERENCES

(1) Assis-Ribas, T.; Forni, M. F.; Winnischofer, S. M. B.; Sogayar, M. C.; Trombetta-Lima, M. Extracellular matrix dynamics during mesenchymal stem cells differentiation. Dev. Biol. 2018, 437 (2), 63−74. (2) Sharma, R. R.; Pollock, K.; Hubel, A.; McKenna, D. Mesenchymal stem or stromal cells: a review of clinical applications and manufacturing practices. Transfusion 2014, 54 (5), 1418−37. (3) Alpizar, A.; Marino, F.; Ramos-Fernandez, A.; Lombardia, M.; Jeko, A.; Pazos, F.; Paradela, A.; Santiago, C.; Heck, A. J.; Marcilla, M. A Molecular Basis for the Presentation of Phosphorylated Peptides by HLA-B Antigens. Mol. Cell. Proteomics 2017, 16 (2), 181−193. (4) Ramos-Fernandez, A.; Paradela, A.; Navajas, R.; Albar, J. P. Generalized method for probability-based peptide and protein identification from tandem mass spectrometry data and sequence database searching. Mol. Cell. Proteomics 2008, 7 (9), 1748−54. (5) Koenig, T.; Menze, B. H.; Kirchner, M.; Monigatti, F.; Parker, K. C.; Patterson, T.; Steen, J. J.; Hamprecht, F. A.; Steen, H. Robust prediction of the MASCOT score for an improved quality assessment in mass spectrometric proteomics. J. Proteome Res. 2008, 7 (9), 3708− 17. (6) Craig, R.; Beavis, R. C. TANDEM: matching proteins with tandem mass spectra. Bioinformatics 2004, 20 (9), 1466−7. (7) Geer, L. Y.; Markey, S. P.; Kowalak, J. A.; Wagner, L.; Xu, M.; Maynard, D. M.; Yang, X.; Shi, W.; Bryant, S. H. Open mass spectrometry search algorithm. J. Proteome Res. 2004, 3 (5), 958−64. (8) Tabb, D. L.; Fernando, C. G.; Chambers, M. C. MyriMatch: highly accurate tandem mass spectral peptide identification by multivariate hypergeometric analysis. J. Proteome Res. 2007, 6 (2), 654−61.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jproteome.8b00384. Supporting Information S1. Materials and methods. Supporting Table S1. Peptide−spectrum matches corresponding to peptides derived from missing proteins. Supporting Table S2. Status in Peptide Atlas of the one-hit wonders reported in this work. Supporting Figure S1. Annotated MS/MS spectra of the tryptic peptides mapping to hyaluronan synthase 1. Supporting C

DOI: 10.1021/acs.jproteome.8b00384 J. Proteome Res. XXXX, XXX, XXX−XXX

Letter

Journal of Proteome Research (9) Schaeffer, M.; Gateau, A.; Teixeira, D.; Michel, P. A.; ZahnZabal, M.; Lane, L. The neXtProt peptide uniqueness checker: a tool for the proteomics community. Bioinformatics 2017, 33 (21), 3471− 3472. (10) Deutsch, E. W.; Overall, C. M.; Van Eyk, J. E.; Baker, M. S.; Paik, Y. K.; Weintraub, S. T.; Lane, L.; Martens, L.; Vandenbrouck, Y.; Kusebauch, U.; Hancock, W. S.; Hermjakob, H.; Aebersold, R.; Moritz, R. L.; Omenn, G. S. Human Proteome Project Mass Spectrometry Data Interpretation Guidelines 2.1. J. Proteome Res. 2016, 15 (11), 3961−3970. (11) Itano, N.; Sawai, T.; Yoshida, M.; Lenas, P.; Yamada, Y.; Imagawa, M.; Shinomura, T.; Hamaguchi, M.; Yoshida, Y.; Ohnuki, Y.; Miyauchi, S.; Spicer, A. P.; McDonald, J. A.; Kimata, K. Three isoforms of mammalian hyaluronan synthases have distinct enzymatic properties. J. Biol. Chem. 1999, 274 (35), 25085−92. (12) Qu, C.; Rilla, K.; Tammi, R.; Tammi, M.; Kroger, H.; Lammi, M. J. Extensive CD44-dependent hyaluronan coats on human bone marrow-derived mesenchymal stem cells produced by hyaluronan synthases HAS1, HAS2 and HAS3. Int. J. Biochem. Cell Biol. 2014, 48, 45−54.

D

DOI: 10.1021/acs.jproteome.8b00384 J. Proteome Res. XXXX, XXX, XXX−XXX